I spent March 19 at DGI's E-Discovery, Records, and Information Management Conference, where AIIM is a proud partner. Three themes kept surfacing across sessions: the traditional boundaries between our fields are disappearing, governance needs to shift from compliance to value, and data is at the center of everything we're trying to do with AI.
Timothy Kootz, Deputy Assistant Secretary for Shared Knowledge Services at the Department of State, was direct about this. "At the end of the day, data is an agency asset." The debate between data management and information management is over. What matters now is working with peers across data management, IT, and data analytics teams.
Angela Kovach reinforced this in her session on connecting e-discovery, information governance, and AI. She pointed out that many agencies still don't know who owns the data or who's driving the thinking around it. The organizations making progress are the ones where CDO, CAIO, and CIO work together instead of protecting territory.
Dan Feith put it even more plainly: "Your IM project is an AI project." The old information lifecycle of capture, manage, deliver is giving way to something more fluid. Mission and needs fuel decisions around centralized data sources that are shared as needed. The focus is shifting from search to analysis, from finding information to making meaning from it.
"Governance isn't the roadblock, it's the roadmap," Kovach said. But only 7% of agencies have documented data guidance for AI. That's not just a compliance gap. It's a mission failure.
Kootz was blunt about the compliance trap: "Compliance without speed or measurable success fails." What matters is outcomes.
"Mission accomplishment matters above all else," Kootz said. Policy should be an enabler and not an obstacle.
Kovach shared stats that are familiar to anyone in our field. 80% of agency data is unstructured with 55-65% growth per year. 33% say AI projects are failing because of unstructured data quality. 80% of legal spend is related to document review.
"There is one collective upstream failure: the data," she said.
Feith called records "the new gold." Centralization of data is key, not just for compliance but to convey value to customers. In the public sector, that means citizens.
While the volume and quality of unstructured data can make it seem like an insurmountable obstacle, there are ways to incrementally improve information management. Kovach outlined a refreshingly practical approach to data and information management. Identify your top three ungoverned data sources. Connect teams around a shared data inventory. Talk to vendors about managing data in place.
Kootz challenged what he calls the "cannot mindset." The organizations that will succeed with AI are those that are able to break the "cannot mindset" and focus on outcomes over compliance. In association management, we'd call this a scarcity mentality. The belief that because we lack staff or funding, we can't invest in growth.
Kootz explained how to break this mindset:
This made me realize we've been asking the wrong question. Information professionals (and AIIM) have been asking whether our organizations are ready for AI. Maybe we should instead be asking whether we are ready to stop saying "cannot" and start asking "how can we?"